[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Idea...
- Subject: Re: Idea...
- From: Cyrus Roton <croton@ridgecrest.ca.us>
- Date: 1995/06/10
- Newsgroups: comp.sys.apple2
- Organization: RidgeNet - SLIP/PPP Internet, Ridgecrest, CA. (619) 371-3501
- References: <3r5vrt$pel@acme.freenet.columbus.oh.us> <Pine.SUN.3.91.950608074311.29941A-100000@owens> <3rbtpi$n8b@gap.cco.caltech.edu>
- Sender: usenet@ridgecrest.ca.us (Ridgenet Usenet admin)
On 10 Jun 1995, Todd P. Whitesel wrote:
>
> SCSI actually allows eight devices -- one per bit of the data bus, which
> is used to assert requests during the arbitration phase. On a standard
> SCSI implementation, the host machine must be counted as a device, hence
> seven other devices may be added.
I think I said eight (zero to seven: that makes eight). I was wrong about
the three device code lines. Actually, the device number switch, on the
back of the box, connects one of the lines to any one of the data lines.
A pulse on that (SEL ??) line is routed to the selected data line for
return to the host. Never the less, another line would be required if you
wanted more than eight codes.
>
> what bits Mhz speed
> ---- ---- --- -----
> Apple bus 8 1 1.00 MB/s
> Ethernet 1 10 1.25 MB/s
>
> Who's losing speed?
>
By "MB/s" do you mean bits/sec or bytes/sec? With serial data flow 100,000
Kilo Bytes per sec equates to 1 Meg Bits per sec. (assuming one start
bit, eight data bits, one stop bit).
What I said about speed loss is sometimes true: If you convert parallel
data to serial data, the speed is cut to one tenth (approx), assuming
that the output clock is the same speed as the input clock. I failed to
recognize that the converter may output data with a higher clock rate than
that on the parallel input side. There are cases where the data
throughput rate on the network is faster than the ability of the
individual computer to supply it.
Thanks for the corrections. Nobody is perfect (especially not I), but, we
can learn from our mistakes.
Cy Roton
- References:
- Idea...
- From: dalloff@freenet.columbus.oh.us (Dave Althoff)
- Re: Idea...
- From: Cyrus Roton <croton@ridgecrest.ca.us>
- Re: Idea...
- From: toddpw@bradbert.ugcs.caltech.edu (Todd P. Whitesel)